PredictNLS

PredictNLS predicts nuclear localization signals (NLSs) in eukaryotic proteins and analyzes their overlap with DNA-binding regions to inform protein subcellular localization and function.

Key Features:

• Extensive database of NLS motifs: The curated dataset includes 91 experimentally verified NLS motifs expanded to 214 potential NLSs through in silico mutagenesis.
• High specificity and sensitivity: PredictNLS matches known nuclear proteins in 43% of cases while producing no matches to non-nuclear proteins.
• Estimation of nuclear import prevalence: The tool estimates that over 17% of all eukaryotic proteins may be imported into the nucleus.
• Overlap with DNA-binding regions: For 90% of proteins with both annotations, NLSs overlap DNA-binding regions, indicating frequent co-localization of these features.
• De novo prediction of DNA-binding regions: Of the 214 motifs, 56 overlap DNA-binding regions and enable prediction of partial DNA-binding regions for approximately 800 proteins in human, fly, worm, and yeast.

Scientific Applications:

• Protein localization studies: Identification of NLSs in uncharacterized proteins to support analyses of protein trafficking and nuclear compartmentalization.
• Functional genomics: Prediction of NLSs and associated DNA-binding regions to aid studies of gene regulation mechanisms.
• Evolutionary biology: Investigation of the evolutionary relationship between DNA-binding domains and nuclear localization via analysis of NLS–DNA-binding overlap.

Methodology:

Starting from a curated set of known NLSs, PredictNLS applies in silico mutagenesis to expand the motif dataset to 214 candidate NLSs.